Psycholinguistic variables in visual word recognition and pronunciation of European Portuguese words: a mega-study approach

ABSTRACT

An increasing number of psycholinguistic studies have adopted a megastudy approach to explore the role that different variables play in the speed and/or accuracy with which words are recognised and/or pronounced in different languages. However, despite evidence for deep and shallow orthographies, little is known about the role that several orthographic, phonological and semantic variables play in visual word recognition and word production of words from intermediate-depth languages, as European Portuguese (EP). The current study aimed to overcome this gap, by collecting lexical decision and naming data for a large pool of words selected to closely represent the diversity of the EP language. Results from multiple regression analyses conducted on the latency data from both tasks place EP in-between the results previously observed in other deep- and shallow-orthographies. These findings indicate that EP represents a pivotal language to study the universality of the processes/mechanisms involved in skilled reading across languages.

KEYWORDS:

• European Portuguese
• megastudy
• psycholinguistic variables
• lexical decision
• naming

Acknowledgements

This study was conducted at Psychology Research Centre (PSI/01662), University of Minho, and supported by the Portuguese Foundation for Science and Technology and the Portuguese Ministry of Science, Technology and Higher Education through national funds, and co-financed by FEDER through COMPETE2020 under the PT2020 Partnership Agreement (POCI-01-0145-FEDER-007653). It is also part of the research project “Procura Palavras (P-Pal): A software program for deriving objective and subjective psycholinguistic indices for European Portuguese words” (PTDC/PSI-PCO/104679/2008).

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Ana P. Pinheiro http://orcid.org/0000-0002-7981-3682

Notes

1. Even though we have not used repeated blocks aiming to directly test practice effects in participants’ performance as Keuleers et al. (2010), the use of lexically similar experimental blocks (see Materials section) allowed us to examine practice effects across blocks even considering that they relied on a different subset of stimuli.

2. Even though the number of phonemes accounted for 0.4% more variance in the latency data from TDL latencies, we opted to use the number of letters in the analyses since this word length measure has been used in most large-scale studies conducted so far (e.g., Ferrand et al., 2010, 2017; New et al., 2006), hence allowing a direct comparison of the results.

3. Note, that if the standard orthographic (ON; Coltheart et al., 1977) and phonological (PN; Luce & Pisoni, 1998) neighbourhood measures were used instead, larger word length effects were observed in both tasks. Specifically, in the latency data from NAM, the percentage of unique variance increased to 8% for ON, F(20, 1910) = 176.98, p < .001, and to 8.3% for PN, F(20, 1910) = 177.993, p < .001, and in the latency data from LDT, to 4.2% both for ON, F(20, 1910) = 70.316, p < .001, and PN, F(20, 1910) = 70.389, p < .001 neighbourhood measures. In the accuracy data from NAM, the percentage of unique variance increased to 0.8% both for ON, F(20, 1910) = 3.779, p < .001, and PN, F(20, 1910) = 3.692, p < .001, although in the accuracy data from LDT it remained the same (5.2%) for PN, F(20, 1910) = 20.545, p < .001, and decreased slightly (4.7%) for ON, F(20, 1910) = 20.566, p < .001.

4. Although Ernestus and Cutler (2015), Goh et al. (2016) and Ferrand et al. (2017) used orthographic and phonological uniqueness point measures in their megastudies, here we only considered the orthographic uniqueness point measure since phonological uniqueness point measures are not available from the P-PAL database (see Soares, Iriarte, et al., 2018).